Engaging citizens to collect daily food prices in rural Indonesia – a proof of concept project
Aug 31, 2015
Pulse Lab Jakarta has just wrapped up the first phase of a food security monitoring deployment in collaboration with Premise and in close cooperation with World Food Programme & Food and Agriculture Organisation. Pulse Lab Jakarta (PLJ), (an innovation initiative of the United Nations Global Pulse and the Government of Indonesia) collaborated with technology company Premise to see whether data collection by citizens in rural parts of Indonesia that is uploaded through technology to Premise’s data platform would work.
This is part of PLJ’s mandate of exploring data innovation methods in Indonesia with the generous support of the Government of Australia. PLJ is keen to use cutting edge information technology and analysis to improve the quality and effectiveness of Indonesia’s development partnerships and saw this approach as a potentially exciting new form of data collection in Indonesia. But would it work in hard to reach and less populated areas of Indonesia?
The overwhelming response was yes! In close consultation with the Food Security Office in Nusa Tenggara Barat (West Nusa Tenggara, a rural province in Eastern Indonesia) on the selection of commodities to be monitored, an index was constructed based on local consumption and expenditure patterns. Twenty commodities were selected and by leveraging Premise’s mobile technology and data science infrastructure, this phase of work was designed to evaluate the degree to which the Premise platform can be used as an effective – and rapid – means to monitor a basket of core consumer food staples in traditional markets in rural and difficult to reach locations.
Data was collected using a local network of more than 200 contributors who were recruited virtually and guided by an app downloaded to their smartphones. Within 10 weeks, more than 65,000 structured observations have been collected and the platform was analyzing price trends across the NTB province on products ranging from tofu and tempeh to eggs and mackerel.
Some of the best takeaways from this proof of concept project were:
- Network recruitment and ramp up time only took one month; within six weeks, the team had achieved broad geographic and product representation across the NTB province. All of this was done virtually.
- Data volumes climbed an average of 40% each week from end of February to mid April 2015 ultimately reaching a peak of close to 12,000 observations per week.
- Prices were captured in over 2,650 unique food trading outlets across Lombok Island and Sumbawa Island. The data collection was not limited to Lombok Island but was also possible in the sparsely populated Sumbawa Island.
Given the ubiquitous nature of mobile phones in Indonesia, it is possibly not so surprising that this proof of concept worked so well but it was still surprising to note the penetrations of smartphones in rural Indonesia and the willingness of the people to capture data on a regular basis (albeit for a small payment!).
PLJ, WFP and FAO were all thrilled with the results from this proof-of-concept project and intend to work together with Premise on the next phase which envisions re-focusing from macro-level trend monitoring to a more structured assessment of ongoing price and scarcity trends at individual markets, the level of granularity necessary for targeting specific interventions.